The files in this repo should allow you to install a working demo for the paper dreams project.
It is highly advised to create a conda environment to run this project successfully.
Run the following commands
conda create --name paperdreams --file requirements.txt
Before continuing, we need to install manually 'en' from the spacy package by running this command in the terminal
python -m spacy download en
On Windows, we encountered an issue creating the 'en' link for spacy. To fix, replace the line nlp = spacy.load('en', disable=['tagger', 'ner', 'textcat']) in text2relations.py with nlp = spacy.load('en_core_web_sm', disable=['tagger', 'ner', 'textcat'])
First we want to grab the svg data files for the drawing assistant functunality.
Grab this folder https://drive.google.com/open?id=1848qA3FOfGm-uKVCG9nVIMTFm2mo6xdB
and put it inside the static > images.
Once that is done, download the GAN models from here https://drive.google.com/open?id=1IyNQ3L67FwH2D3A3f3O-_Z_TnAVLynec
and add the checkpoints folder to the pix2pixHD folder
In the case that you are trying to run the web app on monitors that have high-resolution display (Over 1920x1080) you might want to modify these lines of code 8025 . you can start by removing the division by two
paperdreams_min/static/js/downloadandcolor.js
Line 207 in a22ba7f
paperdreams_min/static/js/downloadandcolor.js
Line 212 in a22ba7f
from the root directory, run the command python __init__.py run
- Guillermo Bernal - Initial work -
- Lily Zhou
- Erica Yuen
- Haripriya Mehta
- Gloria Lin
This project is licensed under the MIT License - see the LICENSE.md file for details
- Danny and Allan Gelman -- explorations for use cases and 3d applications